INTEGRATIVE COMPUTATIONAL Moritz Dörstelmann DESIGN...
Transcript of INTEGRATIVE COMPUTATIONAL Moritz Dörstelmann DESIGN...
219
INTEGRATIVE COMPUTATIONAL DESIGN METHODOLOGIES FOR MODULAR ARCHITECTURAL FIBER COMPOSITE MORPHOLOGIES
Moritz DörstelmannStefana Parascho Marshall PradoAchim MengesJan Knippers University of Stuttgart
A. Custom developed fabrication setup for core-less filament winding with collaborating robots; B. Resulting demonstrator structure: ICD/ITKE Research Pavilion 2013-14
ABSTRACT
How can design computation work as an interface between the virtual design space and the
physical realization space, while forming a point of confluence within a multidisciplinary design
and construction methodology? In an integrated design process for fiber composite structures in
architecture (one based on morphogenesis of fibrous structures in nature), form generation and
materialization are highly interrelated thereby leading to a synergy of form and materiality. This
paper examines the framework of integrative computational design methodologies incorporating
material, structure, fabrication and morphogenetic principles for the design development and dig-
ital fabrication of lightweight fiber-reinforced composite components. This process is discussed
through the case study, ICD/ITKE Research Pavilion 2013-14, including project-specific applica-
tions and the implementation of computational tools.
1
220FABRICATION AGENCY ACADIA 2014 DESIGN AGENCY
INTRODUCTION
Natural morphogenesis is characterized by the capacity of an or-
ganism to develop a form in response to multiple, often contradic-
tory intrinsic and extrinsic stimuli. The resultant form synthesizes
these conditions into one of many possible robust, adaptive out-
comes rather than a unique optimization solution (Webster 1996).
Classical architectural planning processes work entirely different,
as they are often performed as a linear sequence of architectural
design, engineering and implementation. This can either lead to
discrepancies between initial design intentions and build results
or requires multiple iterations of the linear process. Integrative
design methodologies seek to establish processes in architecture
which bear resemblance to natural morphogenesis, where intrin-
sic material and fabrication logics are generative design drivers
rather than inert receptors of form. This creates a bottom-up mor-
phogenetic design approach, synthesizing material and fabrication
constraints through the development of integrative computational
design tools.
In recent years several developments in material, fabrication and
computation have made this methodology possible. Composite
materials, such as fiber-reinforced polymers (FRP), have been
used in performance based, engineering application since they
were developed in the 1930s. These performative materials
have found their way into architectural applications due to a high
strength to weight ratio (Voigt 2007). The formability of FRP pro-
vides a wider spectrum of geometric freedom than that of stan-
dard construction materials (such as plywood sheets or timber
framing), allowing the system to adapt more effortlessly to various
design criteria. This is especially important in an integrative design
process, where form is not predefined as in traditional top-down
design methodologies. Instead, highly differentiated global and
local morphologies are generated as a result of multiple design
criteria. Fiber-reinforced composites provide the opportunity to
determine material behavior through the controlled fabrication
of heterogeneous, anisotropic fiber arrangements. This allows
design processes to expand beyond the traditional scope of archi-
tectural design incorporating the scale of material organization.
Fabrication processes for geometrically complex and performa-
tive composite structures have evolved beyond labor intensive
hand-laying methods to the use of numerically controlled winding
tools. This allows seamless flow of information from complex
computationally generated fiber layouts into precisely fabricated
fiber composite parts. Industrial robotic arms, with their multiple
degrees of freedom and controlled precision, are well-suited
for the task of winding complex geometries. These “generic
machines” are adaptable to a wide range of potential operations
which become specialized through various tools, effectors or
tasks (Menges 2012). The versatility of these machines makes fabri-
cation techniques possible that were previously prohibitive due to
costs or complexity.
The intricacy of natural morphogenesis, which inherently deals
with complex interrelated systems, can now be explored architec-
turally through developing computational design techniques. This
equips architects and designers with the tools to negotiate large
data sets of information, propelling digital explorations into previ-
ously unknown territory (Terzidis 2003). Data-embodied architecture
is, however, not a new concept in design. The notion of building
information modeling (BIM) has existed for several decades and is
a popular methodology in the architecture industry. A BIM model
serves as a virtual database tool for storing and accessing of build-
ing information. BIM is a streamlined building delivery system capa-
ble of coordinating large building data sets. Yet it remains based on
a traditional top-down approach to design, demanding the supply
of information (such as building form) during the design process.
In digital morphogenesis, this initial information is unknown, thus a
bottom-up computational design process is required.
Research at the ICD/ITKE has explored several integrated design
methodologies through the design and fabrication of full scale
architectural demonstrators. In the ICD/ITKE Research Pavilion
2012, a single monocoque shell was developed, that integrated
the structural capacity of FRP with the fabrication constraints
of a 7-axis robotic setup including a rotational positioner. The
formwork was minimized to allow the wrapped fibers to interact
in a core-less winding process (Reichert et al. 2014). The wrapped
geometry was therefore a direct result of the relationship between
the material behavior and the machinic morphospace (Menges 2012).
The ICD/ITKE Research Pavilion 2013-14 continues this research
on integrated design and fabrication processes while focusing on
the development of a modular FRP structural system for a proto-
typical pavilion (Figure 1B). This requires the synthesis of a new set
of design constraints and the deployment of novel design strate-
gies. This paper focuses on the integrated design methodology
that was developed for the ICD/ITKE Research Pavilion 2013-14,
including the integration of the morphological rules derived from
biomimetic principles, material characteristics for the core-less
winding of FRP building components, robotic fabrication limita-
tions, structural performance, architectural framework and orga-
nizational constraints into a single generative process, creating a
continuous and reciprocal exchange of information.
221
Generic integrative design process diagram
the conceptualization of a structural system. The integration of
structural design into a multidisciplinary design process through
novel simulation tools permits engineers and architects to
explore structural systems beyond predetermined typologies,
as well as enabling them to include precise and meaningful
structural analysis into the formally decoupled design process
(Knippers 2013).
Recent developments of computational and fabrication pro-
cesses allow for the generation and materialization of complex
information driven geometries. This opens up greater potentials
for the transfer of functional principles from biology to novel
material and structural systems within a multidisciplinary pro-
cess, while generating potential biological insight through the
use of cross-disciplinary methodologies within a reverse biomi-
metic process. The convergence of material and materialization
allows fabrication technology and material behavior to push
new architectural possibilities as creative inputs.
2
INTEGRATIVE COMPUTATIONAL DESIGN METHODOLOGIESDÖRSTELMANN, ET. AL.
INTEGRATIVE DESIGN COMPUTATION WITHIN A MULTIDISCIPLINARY DESIGN PROCESS
In order to orchestrate the complex network of interrelations
within a multidisciplinary design and construction process, design
computation is used to translate various process inputs into tool
parameters and encoding the interrelations into an algorithmic
framework (Figure 2). Setting up a robust tool is not merely to digi-
tize or automate the design process (Bechthold 2010), but allows for
the production and evaluation of multiple design iterations which
can be used to navigate the design space in an explorative way.
This is an excellent usage of digital design processes, where the
computational capacity to negotiate complex interrelationships
works as a tool for the interface and expansion of design thinking.
By serving as a point of confluence computational design meth-
odologies allow an interdisciplinary information exchange which
drastically changes the way each field operates.
Traditionally, structural engineers have been trained to identify and
think in typologies of structures. This was necessary to identify
the right set of formulas for calculation or to provide guidance in
FABRICATION AGENCY 222ACADIA 2014 DESIGN AGENCY
BIOMIMETICS
Functional morphologies and material organization processes in
biology are highly relevant within an architectural context. They
achieve fundamental architectural tasks like materialization of
form and functional integration in a materially efficient manner.
As part of the ICD/ITKE Research Pavilion 2013-14, various natu-
ral fiber composite structures were investigated. The structural
morphology of beetle elytra, particularly those in flying beetles,
served as a versatile role model for the abstraction of light weight
construction principles, since it is both the rigid protection for the
beetle and has a minimized weight for flight (Figure 3A). The struc-
tural capacity of the elytra is based on the complex geometry and
anisotropic organization of the natural chitin fiber composite ma-
terial. From this biological role model the double layered shell with
internal hyperbolically shaped interconnections (Trabeculae, Figure
3B), was abstracted into a modular arrangement of components.
Both the structural morphology of the elytra and the distribution of
trabeculae provided guiding principles for efficient material usage
and geometric differentiation of FRP building components.
MATERIAL AND STRUCTURE
In an integrated design methodology materiality is regarded as an
intrinsic part of the design process, rather than playing a subordi-
nated role to a predetermined geometric form. Subsequently, the
relationship between material behavior and the form generation
process results in a design outcome which is strongly informed
by material potentials and constraints.
While topologically and conceptually informed by the biomimetic
investigation, the specific materialized geometries of the compo-
nents are defined by the core-less filament winding process, a
fabrication system that requires no positive mold for the genera-
tion of double-curved geometries (Waimer 2013) and the winding
syntax, the systematic sequence of fiber placement. The structur-
ally and materially informed winding syntax is intended to create
adequate fiber-fiber interaction in order to achieve a structurally
active composite surface.
Several winding syntax codes were developed in order to cover
the various material functions of the fiber-based system. The
winding syntax of the initial glass fiber layer describes heli-
coidal fiber arrangements which ensure forming of anticlastic
geometry. This is the result of the system deforming under the
tension applied by the subsequent layers of fibers. There is an
Leptinotarsa decemlineata: A. Top view, light microscopy by Dr. Thomas van de Kamp; B. Section view into elytron’s internal trabecel bracing struc-ture, SEM by Prof. Oliver Betz
interdependency between the geometric variation of each com-
ponent and the winding syntax, which must be resolved to trans-
late the differentiated geometry into the hyperbolic column-like
structures. The glass fiber layer acts as scaffolding for the car-
bon fiber layers, which structurally reinforces each component
and the global geometry. For this purpose, two winding syntaxes
were developed. The first is a layer of carbon fiber that transfers
generic, omni-directional structural forces in each component;
the second reinforces the components in regard to predominant
global structural force flows.
Structural input occurs through all hierarchical levels, influencing
material placement and properties, components local geometries
and the global design system. The process is accompanied by
structural FE simulations (Figure 5), which lead to results that were
then interpreted into parameters for the design process. The
differentiated carbon fiber layer integrates both geometric require-
ments and variable boundary conditions such as possible com-
ponent to component connection points and structural analysis
information. This enables the position and direction of the carbon
fiber to be differentiated according to the load distribution along
the global structure, and allows for the effective transfer of loads
through adjoining components. Through simulating the global
structure, high forces occurring in the system are identified and
translated into fiber directions. This results in carbon fibers which
are specifically placed according to the force vector orientations
and fiber density that is defined by the amplitude of the vector
3
Carbon and glass fiber building component: A. Front view; B. Differentiated carbon fiber reinforcement; C. Omni-directional carbon fiber reinforcement
4
223
(Figure 6). Both carbon fiber layers are strongly informed by the
geometric constraints of the glass fibers, since they must ne-
gotiate the double curved surface geometry and ensure a proper
fiber-to-fiber bond on each component. The algorithm used to cre-
ate the adaptive winding syntaxes bridges the gap between the
structural simulation and the geometric environment, while being
informed by the other conditioning factors (the global constraints
and fabrication requirements).
MATERIALIZATION
Bottom-up design strategies often result in complex geometries
which explore the potentials of an open-ended system. This pos-
es a difficult challenge for implementation as standard methods
of fabrication are not always suitable. In an integrated design
process, in which fabrication is intrinsically linked, materialization
is inseparable from material and form. This requires the develop-
ment of novel fabrication strategies in parallel to computational
techniques in order to negotiate the expression of form and
materialization. The geometric outcome is then charged with fab-
rication constraints, and therefore limiting what is buildable to the
machinic morphospace (Menges 2012). In the ICD/ITKE Research
Pavilion 2013-14, the fabrication strategy is closely tied to the
component and system geometry as well as the fiber syntax. The
general size and complexity of the components determine the
approximate fabrication setup needed, for example, two robots,
their distance from each other and the size of their effectors
(Prado 2014) (Figure1A). This setup, in turn, plays a crucial role in
determining the limitations of the components geometry such as
the degree of non-planarity, maximum component size and angle
between components, thus defining the design domain of the
overall project (Figure7). The fiber syntax, which was hand-modeled
with a scaled version of the fabrication process, was incrementally
optimized to streamline winding. The fiber syntax also formed the
basis for the robotic winding code which was generated from the
integrated data using custom scripted computational processes.
Computation played a vital role in integrating fabrication into the
design process. Not only does it create opportunities to mitigate
design and fabrication challenges, but also allows for the coordi-
nation of the digital model and fabrication setup. Custom scripts
were developed to translate component geometry from the
digital information model to the robotic assembly code for each
effector frame. The robot assisted assembly of the effectors was
not only intended to automate the process, but to compensate
for manual inaccuracies through robotic precision. Digitally fab-
ricated guides are used to secure the effector frame geometry,
which has a high degree of geometric freedom, to the correct
orientation for fabrication. Defining the geometry of these guides
INTEGRATIVE COMPUTATIONAL DESIGN METHODOLOGIESDÖRSTELMANN, ET. AL.
Finite element analysis of the global structural system5
Structurally differentiated winding syntax development diagram: A. Finite Element simulation; B. Force vectors; C. Traced force directions; D. Differentiated Fiber reinforcements
6
FABRICATION AGENCY 224ACADIA 2014 DESIGN AGENCY
required the integration of multiple, potentially conflicting, fab-
rication criteria (attachment points, angles of rotation, support
planarity and fiber interference). This digital design tool allowed
for the control of multiple parameters and the visual feedback of
underperforming criteria for quick evaluation and iteration.
Since the interrelationship between the robotics setup, winding
syntax and component geometry all jointly contributed to the
fabrication limitations, it becomes increasingly important to
optimize the fabrication process and simulate potential prob-
lems virtually in order to utilize the extents of the fabrication
setups geometric possibilities. Genetic algorithms are used to
determine the tool planes for each effector and base rotation
for winding. These cannot be determined geometrically from
the component center point or column axis, but rather require
cross-referencing all the robot orientation planes with the robot-
ic limitations of the twelve axis setup in order to determine if any
orientations are out of reach. A robotic simulation and adaptive
base plane are used to avoid collisions. These computational
processes demonstrate the dependency between form genera-
tion and fabrication as well as the need for integrative tools for
the design of both.
SOLUTION SPACE
The design space encompasses the domain of all possible out-
comes which can be generated by computational design tools.
To achieve meaningful results within a certain target zone, a
well-defined design domain is required.
Component morphologic parameters and Global system variations7
Within an integrated computational design process, the design
space is an outcome of the reciprocal constraints between the
various integrated process parameters. It serves as a purposeful
truncation of the design outcomes. Over- or under-constraining
of the solution space can either lead to predefined results or
unfeasible outcomes.
To maintain the explorative character of the computational
design process and diversity of results while still generating
feasible designs, the design space needs to serve as a boundary
within which manifold solutions can emerge through the inter-
play of the various integrated process parameters.
In the ICD/ITKE Research Pavilion 2013-14, a project-specific solu-
tion space is defined in order to maintain a feasible scope of the
project for full scale implementation. Therefore basic boundary
conditions regarding size and covered area, as well as the overall
spatial layout, are defined to serve as the architectural domain for
the bottom up computational design process. Based on these
boundary conditions, an initial network of component arrange-
ments is determined by principles regarding the distribution of
trabeculae which is found in various beetle elytra. This initial com-
ponent layout served as a starting point for the more complex
negotiation of other process parameters. The integration of fabri-
cation constraints, such as the general range in size of producible
components and geometric rules for component distribution
based on biomimetic principles, led to meaningful constraints of
the solution space for the computational design process.
225 INTEGRATIVE COMPUTATIONAL DESIGN METHODOLOGIESDÖRSTELMANN, ET. AL.
The computational strategy defining the initial component layout
is based on interaction between the boundary conditions utilized
in the architectural framework and a bottom-up behavioral integra-
tion method. The component layout emerges through an iterative
process of cellular arrangement, which is based on simple rules for
neighboring cell interactions (Figure 8). Biomimetic design principles
and fabrication processes are integrated into the computational
design process from the beginning, thereby informing the initial cel-
lular layout and contributing to the definition of the design space.
NEGOTIATING INTERRELATIONS
Within the design space defined by this initial component to-
pology, a morphogenetic design process unfolds which instru-
mentalizes process constraints as active drivers for variation and
adaptation of structural morphologies on a global and component
level. The integrative design methodology does not rely on linear
optimization strategies, but utilizes a flexible design exploration
tool that navigates through the solution space by negotiating
potentially antagonistic constraints. The system resolves itself
through a non-linear open-ended process into one of many states
of equilibrium (Attar 2010).
In order to establish the interrelationships between different fields,
the influences of each must be represented systematically. A phys-
ics simulation forms the framework for the synthesis of both phys-
ical forces (such as gravity) and non-physical “forces” (fabrication
constraints) (Piker 2013). The components morphology is represent-
ed by particles connected with springs. Forces are employed as
abstract representations of extrinsic and intrinsic process require-
ments. This allows the components morphology to adjust within a
cyclical interaction of a particles spring system ensuring the emer-
gence of a global morphology through local interactions.
Information gathered during hand and robotic test winding was
formulated as geometric rules to ensure proper fiber-fiber inter-
action, and encoded into forces or constraints acting onto the
spring system. This includes component polygon angles, edge
length relations or the overall components diameter to height ratio
(Figure 7). Other forces are informed by constraints derived from
the specific robotic setup and certain effector characteristics in
order to avoid out of reach errors or collisions within the wind-
ing process. This applies for example to the allowable range of
non-planarity for each component polygon and the overall compo-
nent diameter. The structural principles which were abstracted by
biomimetic investigation of the beetle elytra structural morpholo-
gy were transferred into the design process through the definition
of forces, affecting the components global distribution as well as
their local morphologic differentiation.
Various methods of evaluating criteria are applied to the system
within the process. Geometric relationships are verified, for each
iteration of the computational solver, and are used to activate or
deactivate adaptive parameters. Some forces are controlled by
gradual falloffs or soft transitions, while other evaluation criteria
rely on rigid constraints to override other forces. The differentiated
effects of the design drivers allow for the hierarchical control and
evaluation of each process parameters impact. A finite element
solver calculated information from the structural performance of
each iteration, creating a morphologic response to specific loading
scenarios in order to implement the biological lightweight construc-
tion principles (Figure 9). Evaluation processes are integrated into the
system which allow for the monitoring and adjustment of critical
design or fabrication criteria (component sizes, overall enclosed
volume and area) which lead to an online estimation of required
resources (winding time and material requirements).
CONCLUSION
The design methodology discussed here, through the case
study of the ICD/ITKE Research Pavilion 2013-14, highlights
the potential for design computation to serve as a common
language within a multidisciplinary process (Figure 10). The
non-linear design process allows for parallel bottom up devel-
opments and exploration of solutions which emerge through
the instrumentalization of process constraints as design drivers.
Ultimately, this leads to novel possibilities in design exploration
and materialization of architectural structures.
The project-specific development of a computational design
framework within the ICD/ITKE Research Pavilion 2013-14 led
to the implementation of structural principles derived from bi-
ological role models, and to a parallel bottom up exploration of
Behavior based cell arrangements8
FABRICATION AGENCY 226ACADIA 2014 DESIGN AGENCY
Iterative physics based simulation with integrated FE feedback9
material capacities and fabrication strategies. This enabled the
materialization of an architectural prototype that is built from 36
fiber composite components with individually differentiated fiber
layouts, and adapted morphologies resulting in a double layered,
highly efficient lightweight structure (Figure 11, 12). The complex-
ity in the components geometric articulation and the high level
of detail in its material arrangement showcase how novel ex-
periential qualities arise through the application of novel design
and fabrication processes in architecture. The research pavilion
covers an area of 50.3 m² and encloses a volume of 122 m³ with
an overall weight of 594 kg.
Future challenges include expanding integrative design method-
ologies to include the use of sensor data and real time robotic
control, allowing for design adaptability during the materialization
process. Such an approach would require a shift from instruction
based fabrication, in which robots follow preprogrammed move-
ments, towards flexible fabrication strategies, where movements
can respond to changing fabrication conditions.
The main outcome of the presented research is the development
of the integrated design methodology, whose full potential is yet
to be explored. The inherent capacity of this design methodology
is to generate form as the direct outcome of the underlying pro-
cesses of material organization and functional integration. While
the case study achieved morphological variation of a modular
lightweight construction based on material capacity and fabrica-
tion processes, functional integration should be further explored.
In architectural terms that may include climate adaptation (weath-
er proofing, insulation and thermal regulation); thus, a fully devel-
oped building construction system could be developed through
the integration of further architectural aspects. Project specific integrative design process diagram10
227 INTEGRATIVE COMPUTATIONAL DESIGN METHODOLOGIESDÖRSTELMANN, ET. AL.
ACKNOWLEDGEMENTSThe work presented in this paper is part of a joint project between the University of Stuttgart and the University of Tübingen. The authors would like to express their gratitude towards their fellow investigators: Prof. Oliver Betz, Institute of Evolution and Ecology, Evolutionary Biology of Invertebrates, University of Tübingen; Prof. James Nebelsick, Department of Geosciences, Paleontology of Invertebrates, University of Tübingen; Dr. Thomas van de Kamp; Tomy dos Santos Rolo; Prof. Dr. Tilo Baumbach, ANKA/Institute for Photon Science and Synchrotron Radiation, Karlsruhe Institute of Technology (KIT); Dr.-Ing. Thomas Stehle, Rolf Bauer, Michael Reichersdörfer, Institute for Machine Tools, Universität Stuttgart; Dr. Markus Milwich, Institute of Textile Technology and Process Engineering, ITV Denkendorf. The project was partially funded by: Competence Network Biomimetics, KUKA Roboter GmbH, SGL Group, Sika AG and AFBW—Allianz Faserbasierte Werkstoffe Baden-Württemberg.
REFERENCESAttar, Ramtin, Robert Aish, Jos Stam, Duncan Brinsmead, Alex Tessier, Michael Glueck, and Azam Khan. 2011. “Embedded Rationality: A Unified Simulation Framework for Interactive Form Finding.” International Journal of Architectural Computing. doi:10.1260/1478-0771.8.4.399.
Bechthold, Martin. 2010. “The Return of the Future: A Second Go at Robotic Construction.” Architectural Design 80: 116–121. doi:10.1002/ad.1115.
Menges, Achim. 2013. “Morphospaces of Robotic Fabrication.” In Robotic Fabrication in Architecture, Art and Design: Proceedings of the Robots in Architecture Conference 2012, 28–47.
Piker, Daniel. 2013. “Kangaroo: Form Finding with Computational Physics.” Architectural Design. doi:10.1002/ad.1569.
Prado, Marshall, Moritz Dörstelmann, Tobias Schwinn, Achim Menges, and Jan Knippers. 2014. “Core-Less Filament Winding.” In Robotic Fabrication in Architecture, Art and Design 2014 SE - 19, edited by Wes McGee and Monica Ponce de Leon, 275–289. Springer International Publishing. doi:10.1007/978-3-319-04663-1_19. http://dx.doi.org/10.1007/978-3-319-04663-1_19.
Pavilion: Interior image11a Pavilion: Overhang image11b
Pavilion: Component system12a
Pavilion: Exterior image12b
FABRICATION AGENCY 228ACADIA 2014 DESIGN AGENCY
Reichert, Steffen, Tobias Schwinn, Riccardo La Magna, Frederic Waimer, Jan Knippers, and Achim Menges. 2014. “Fibrous Structures: An Integrative Approach to Design Computation, Simulation and Fabrication for Lightweight, Glass and Carbon Fibre Composite Structures in Architecture Based on Biomimetic Design Principles.” CAD Computer Aided Design 52: 27–39. doi:10.1016/j.cad.2014.02.005.
Terzidis, Kostas. 2003. Expressive Form: A Conceptual Approach to Computational Design. Spon Press.
Knippers, Jan 2013. “From Model Thinking to Process Design.” In Architectural Design 02/2013, Wiley, 74–81.
Waimer, Frederic, Riccardo La Magna and Jan Knippers. 2013. “Integrative Numerical Techniques for Fibre Reinforced Polymers—Forming Process and Analysis of Differentiated Anisotropy.” Journal of the International Association for Shell and Spacial Structures 54(4): 301–309.
Webster, Gerry and Goodwin, Brian C. 1996. Form and Transformation: Generative and Relational Principles in Biology. Cambridge University Press.
Voigt, Pamela. 2007. Die Pionierphase Des Bauens Mit Glasfaserverstärkten Kunststof-fen (GFK) 1942 Bis 1980. Bauhaus-Universität Weimar.
IMAGE CREDITSFigure 1A, 1B, 2. ICD/ITKE University of Stuttgart (2014)
Figure 3A. Dr.Thomas van de Kamp (2013)
Figure 3B. Prof. Oliver Betz, Anne Buhl, University of Tübingen (2013)
Figure 4A, 4B, 4C. ICD/ITKE University of Stuttgart (2014)
Figure 5. ICD/ITKE University of Stuttgart (2014)
Figure 6A–6D. ICD/ITKE University of Stuttgart (2014)
Figure 7–10. ICD/ITKE University of Stuttgart (2014)
Figure 11A and 11B. ICD/ITKE University of Stuttgart (2014)
Figure 12A and 12B. ICD/ITKE University of Stuttgart (2014)
MORITZ DÖRSTELMANN is a Research Associate at the Institute for Computational Design (ICD) at the University of Stuttgart. He started his architecture studies at the RWTH Aachen University in 2005. In 2011 he graduated with distinction from University of Applied Arts in Vienna where he studied in the master class of Zaha Hadid and Patrik Schumacher. The focus of Moritz’s research is the investigation of biological fiber based structures, regarding functional integration, structural performance and material organization strategies, and trans-fer of their underlying principles into architecture by an integrative design processes employing computational design strategies and digital fabrication.
STEFANA PARASCHO is a research assistant at the Institute of Building Structures and Structural Design (ITKE) at the University of Stuttgart. She has worked in DesignToProduction Stuttgart focusing on geometrical problems and construction prin-ciples for complex geometries. She graduated in 2012 from the University of Stuttgart with the Diploma Thesis “Integrative Design Methods for the Development of Freeform Geometries.” She has taught at the Institute of Computational Design (ICD) and worked in Knippers Helbig Advanced Engineering Stuttgart. Her main research interest lies in integrative design methods implemented through computational models, such as Agent-Based Systems or Evolutionary Algorithms.
MARSHALL PRADO is a Research Associate for the Institute for Computational Design at the University of Stuttgart. He holds a Bachelor of Architecture from North Carolina State University and advanced degrees as a Master of Architecture and a Master of Design Studies in Technology from the Harvard University Graduate School of Design. His current research interests include the integra-tion of computation and fabrication techniques into material systems and spatial design strategies.
ACHIM MENGES is a registered architect and professor at the University of Stuttgart where he is the founding director of the Institute for Computational Design. Currently he also is Visiting Professor in Architecture at Harvard University’s Graduate School of Design and Visiting Professor of the Emergent Technologies and Design Graduate Program at the Architectural Association, London.
JAN KNIPPERS specialises in complex parametrical generated structures, as well as the use of innovative materials. Since 2000 Jan Knippers is head of the Institute for Building Structures and Structural Design (itke) at the University of Stuttgart and involved in many re-search projects on fiber-based materials and biomimetics in architec-ture. He is also partner and co-founder of Knippers Helbig Advanced Engineering The focus of their work is on efficient structural design for international and architecturally demanding projects. Jan Knippers completed his studies of engineering at the Technical University of Berlin in 1992 with the award of a PhD.